100%(1)1 out of 1 people found this document helpful
This preview shows page 5 - 7 out of 10 pages.
Sampling Adequacy (KMO) was applied on the data. The results of the Bartlett’s Test of Spherecity and KMO test are shown in table 2. Table 2 : KMO and Bartlett's TestKaiser-Meyer-Olkin Measure of Sampling Adequacy. .784 Bartlett's Test of Sphericity Approx. Chi-Square 3.028E3 df 185 Sig. .000 Table 2 show that Bartlett's Test of Spherecity (approx. Chi-Square is 3.028E3, Degree of Freedom is 185, and significance is 0.000) and Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) value is 0.784 which shows that data is adequate for factor analysis. Factor Analysis Principal Component Analysis by use of Varimax Rotation was done for extracting the underlying factors and to check construct validity. The outcome of Principal component analysis is given in table 3. Table 3: Factor Analysis Construct No. of items Factor loadings Eigen Value Trust 12 0.747-0.833 2.567 Perceived Value 10 0.693-0.798 2.017
International Journal of Advanced Science and Technology Vol. 29, No. 11s, (2020), pp. 975-984ISSN: 2005-4238 IJAST Copyright ⓒ 2020 SERSC980 Positive reviews 8 0.644-0.801 1.446 Extraction method: Principal component analysis Factor loading of 0.5 or more was considered for all the items and Eigen value of more than 1 was considered for extraction of factors (Hair et al., 2010). Overall 3 factors were retained as their Eigen values was greater than 1 and based on category of items in these factors they were named as-Trust, perceived value and positive response.Table 4: Reliability Analysis Variables Cronbach’s AlphaTrust 0.843 Perceived Value 0.807 Positive reviews 0.786 The reliability of the scale was tested by employing the Cronbach’s Alpha reliability statistics. The computed reliability coefficient of Cronbach’s Alphawas more than 0.6 for all measures which implies the high reliability for the measure used (Hair et al., 2010). Multiple regression Analysis To find relationship between consumer buying behaviour (dependent variable) and trust, perceived value and positive reviews (independent variables) regression analysis was done. The general linear regression model used is of the following form: Y =βo+β1x1+β2x2+β3x3+e Where Y is the dependent variable, the β’s are the parameter estimates, the x’s are theIndependent variables, and e is the residual term. Y-variable is consumer’s buying behavior on social media. The independent variables are trust, perceived value and positive response. Table 5: Regression Analysis Independent variables βSE(β) t-value Sig.t (α)R2Model F-value Sig. F Constant 1.643 0.209 7.592 0.000 0.60 47.83 0.000 Trust 0.496 0.049 7.220 0.000 Perceived value 0.543 0.069 14.042 0.000 Positive reviews 0.268 0.039 6.416 0.000 Table 5 shows that the variables trust, perceived value and positive review are all statistically significant predictors (α ≤ 0.05) of consumers’ buying behavior on social media. R2is 0.60, which explains 60 percent of the variance in consumers’ buying behavior on social media. As seen in table 5 all variables i.e. trust, perceived value and positive reviews are positively related to consumers’ buying behavior in social media. Inspection of the βcoefficients shows that perceived value is the most influential explanatory variable at .543, followed by trust at .496 and positive reviews at .268 for consumers’ buying behaviour on social